Healthcare AI Guy Weekly Newsletter | 4/9

Our Picks ✨

Highlights if you’ve only got 2 minutes…

1/

AI 100: The most promising AI startups of 2024

CB Insights just released this year’s AI 100 report — their list of the 100 top private AI companies across the globe. This year’s winners are tackling some of the hardest challenges across industries, from new AI architectures to precision manufacturing. Out of the 100 winners, 7 were healthcare-focused. Those below:

• OpenEvidence: AI-powered clinical decision support system

• Gesund.ai: AI assurance platform in healthcare to bring clinical-grade AI solutions to market

• Bioptimus: AI startup focused on developing foundational models for biology

• Iambic: AI-driven platform for drug discovery

• Genesis Therapeutics: AI-powered drug discovery

• CHARM Therapeutics: 3D deep learning and AI drug discovery tech

• Isomorphic Labs: AI-powered drug discovery

A few fast facts on the top 100 cohort: (1) raised a total of +$28B across 240+ equity deals since 2020 — OpenAI raised over 40% of that total, with $12B. (2) 68% of the list are early-stage or have yet to raise outside equity. (3) 31% of this year’s cohort are based outside the US. In terms of the healthcare names, OpenEvidence has the 7th highest valuation per employee out of the entire list, at $35.4M. (link)

2/

Epic AI validation tool to test, monitor models

The dominant EHR vendor, Epic, plans to release an AI validation software suite to enable healthcare orgs to evaluate AI models at the local level and monitor those systems over time. Epic is launching the validation tool amid a national debate on how to best validate AI and test algorithms in healthcare.

Epic’s SVP of R&D Seth Hain said that the “AI trust and assurance software suite” automates data collection and mapping to provide nearly instant analysis on model performance. Epic is focusing on local testing because of the difference in patient populations. It sounds like we could get a look at the new capabilities in the next four to six weeks, with more updates to come throughout the summer.

Regardless, AI in healthcare could benefit from a widespread AI validation tool and numerous industry leaders see Epic’s move as a step in the right direction. (link)

3/

Who’s liable when an AI system fails?

The American Medical Association (AMA) wants Congress to address concerns of medical malpractice for physicians using AI tools. The debate over AI liability is less about how much compensation a patient gets, but who bears the blame for a mistake.

According to tech firms and some hospitals, physicians making the final call in care are ultimately responsible for their decisions. But, according to a recent NEJM paper from Stanford researchers, we still don’t know who’s liable when an AI system fails. The authors outline how providers are currently putting themselves at risk when implementing AI solutions despite complex models often having billions of variables that make it difficult to find specific flaws to point to in court.

AI developers could shield themselves by seeking “regulatory approval,” but case law suggests doctors aren’t likely to have similar avenues to avoid lawsuits. It’s fair to say hospitals and startups are still in the dark about liability within medical AI, making security and compliance really important for further progress. (link)

Tools & Partnerships 🔧

Latest on business, consumer, and clinical healthcare AI tools and partnerships…

TOOLS

• Kaiser AI system resolves one-third of patient messages: Kaiser Permanente used an AI-powered system to categorize 4.7 million patient messages by topic and forward them to regional team management. The AI-powered system resolved one-third of patient messages, helping clear clinicians’ inboxes. (link)

• Eko Health AI for heart failure detection: Smart stethoscope company Eko Health announced FDA clearance of its Low Ejection Fraction AI algorithm (Low EF AI), which expands heart failure detection to stethoscope exams at primary care offices (vs. specialist-performed echo exams at clinics/hospitals). (link)

• AI-driven test to predict sepsis in hospitals: The healthtech company Prenosis announced its AI-powered diagnostic tool for sepsis is the first to receive approval from the FDA. Sepsis occurs when the body has an extreme response to an infection, and it is often deadly. (link)

• AI develops potential new drugs for antibiotic-resistant bacteria: Stanford Medicine researchers devise a new artificial intelligence model, SyntheMol, which creates recipes for chemists to synthesize the drugs in the lab. (link)

PARTNERSHIPS

• Pager + Google Cloud: Pager Health and Google Cloud announced a trio of new GenAI solutions to enhance payor productivity and member experience as part of their deeper partnership. The solutions include FAQ Bot (answers common health and benefits questions while recommending next best steps), Sentiment Analysis (assists care and service teams in better understanding how member emotion is changing), and Chat Summation (as the name suggests, it summarizes patient-provider chats). (link)

• Aegis Ventures + Nine Health Systems: A digital consortium aimed at developing and deploying digital health solutions has been established, with participation from nine health systems, including Northwell Health and Memorial Hermann Health System. Led by Aegis Ventures, the consortium announced its focus on generating firms that address both clinical and operational challenges in health care. Aegis indicated that the portfolio would prominently feature companies specialized in AI. (link)

Deal Desk 💸 

Spotlight on latest capital raises, M&A, and investments…

FUNDING

• AITRICS, a predictive medical analytics platform, raised $20M in Series B financing. The startup’s platform leverages AI to predict the likelihood of patient health deterioration, using various data types including various vital signs, blood test results, age, and more. (link)

• Manifold, an AI-based clinical research platform to help reduce the administrative burden at cancer centers, raised $15M in series A funding. TQ Ventures led with new investors Calibrate Ventures and SK Ventures also participating. (link)

• Nucleai, a pioneer in spatial AI biomarkers, raised $14M in funding led by M Ventures, the corporate venture arm of Merck KGaA. (link)

• HD, raises $5.6M to build healthcare chatbots for South East Asia. (link)

Other Relevant News 🔍

News, podcasts, blogs, tweets, resources, etc…

• Healthcare sector emphasizes need for AI expertise (link)

• Chief AI Officer a newcomer in some healthcare c-suites (link)

• Analysis: AI helps pharma develop new cures but imperils drug patents (link)

• Drugmakers developing regulatory pathway for AI in clinical trials (link)

• Nabla appoints new clinical director (link)

• Algorithm-based system fails to capture nuances of care for vulnerable seniors (link)

• Oncologist perspectives on AI (link)

Visuals of the Week 📸

Funny memes, cool pics, and interesting data from around the web…